Diabetic retinopathy (DR) is a leading cause of blindness worldwide. Early detection and treatment can prevent blindness, but manual diagnosis is time-consuming and prone to errors. Convolutional neural networks (CNNs) have demonstrated effectiveness in DR detection, with the VGG architecture being a popular choice for such tasks.This project presents a CNN-based DR detection system using the VGG architecture. The system was trained on a publicly available dataset of fundus images, achieving an accuracy of 99.88% on the training set and 93.40% on the test set.
-
Convolutional Neural Networks (CNNs): CNNs are a type of deep learning algorithm commonly used for image classification tasks. They are capable of learning hierarchical features from images, making them suitable for DR detection.
-
VGG Architecture: The VGG architecture, developed by the Visual Geometry Group at Oxford University, is known for its simplicity and high accuracy in various image classification tasks. In this project, the VGG8 architecture was used, which is a variant of the standard VGG model.